Data Location
Runs
Smoothing
Matches ()
Matches ()
Main Graph
Auxiliary Nodes
Functions
Continue...
Over Session Runs:
Till Condition Met by Watched Tensor
Session Runs
| Feeds | Fetches | Targets | #(Devices) | Count |
|---|
""
| Tensor | Count | DType | Shape | Value |
Health Pill
|
|---|
|
|
|
|
|
|
Tensor Values
Enable all (not just sampled) steps. Requires slow disk read.
Step of Health Pills:
Health Pill
Numeric Alerts
Alerts are sorted from top to bottom by increasing timestamp.
| First Offense | Tensor (Device) | Event Counts |
|---|
TPU Compatibility
Run ()
Trace inputs
Color
Graph |
(* = expandable) |
|
Namespace*
?
|
|
|
OpNode
?
|
|
|
Unconnected series*
?
|
|
|
Connected series*
?
|
|
|
Constant
?
|
|
|
Summary
?
|
|
|
Dataflow edge
?
|
|
|
Control dependency edge
?
|
|
|
Reference edge
?
|
BOOKMARKS ()
Open this drawer to save a set of views of the projection, including
selected points. A file containing the bookmarks can then be saved and
later loaded to view them.
DATA
Load data from your computer
Step 1: Load a TSV file of vectors.
Example of 3 vectors with dimension 4:
0.1\t0.2\t0.5\t0.9
0.2\t0.1\t5.0\t0.2
0.4\t0.1\t7.0\t0.8
0.2\t0.1\t5.0\t0.2
0.4\t0.1\t7.0\t0.8
Step 2 (optional): Load a TSV file of metadata.
Example of 3 data points and 2 columns.
Note: If there is more than one column, the first row will be parsed as column labels.
Note: If there is more than one column, the first row will be parsed as column labels.
Pokémon\tSpecies
Wartortle\tTurtle
Venusaur\tSeed
Charmeleon\tFlame
Wartortle\tTurtle
Venusaur\tSeed
Charmeleon\tFlame
Click outside to dismiss.
Publish your embedding visualization and data
If you'd like to share your visualization with the world, follow these simple steps. See this tutorial for more.
Step 1: Make data public
Host tensors, metadata, sprite image, and bookmarks TSV files publicly on the web.
One option is using a github gist. If you choose this approach, make sure to link directly to the raw file.
Step 2: Projector config
Step 3: Host projector config
After you have hosted the projector config JSON file you built above, paste the URL to the config below.Click outside to dismiss.
| Checkpoint: | |
| Metadata: |
/
/
t-SNE
PCA
Custom
Iteration: 0
For fast results, the data will be sampled down to points.
#
Variance (%)
#
Variance (%)
#
Variance (%)
PCA is approximate.
Total variance
3D controls
Rotate Mouse left click.Pan Mouse right click.
Zoom Mouse wheel.
Holding ctrl reverses the mouse clicks.
2D controls
Pan Mouse left click.Zoom Mouse wheel.
Click anywhere to dismiss.
Section 1: Summary of input-pipeline analysis
The overall performance is input-bounded because on average (over all steps) the device has spent % of time (standard deviation = ) waiting for input data [see Section 2 below for details].
Recommendation for next step:
Section 2: Device-side analysis details
Section 2.1: Device step time
|
Device step-time statistics (in ms) Average: ms (σ = ms) Range: - ms |
milliseconds |
training step number |
Section 2.2: Range of device time waiting for input data across cores at each step
|
% of device step time waiting for input data (average over the maximum waiting time across cores at each step) Average: % (σ = %) Range: - % |
% of device step time |
training step number |
Section 3: Host-side analysis details
What can be done to reduce above components of the host input time:
Host side analysis can not be done without instrumentation.
Click the "Show" button below to see the source data of the breakdown.
| Input Op | Count | Total Time (in ms) | Total Time (as % of total input-processing time) | Total Self Time (in ms) | Total Self Time (as % of total input-processing time) | Category |
|---|
Average step time (lower is better): ms (standard deviation = ms)
Host idle time (lower is better):
TPU idle time (lower is better):
Utilization of TPU Matrix Units (higher is better):
milliseconds |
training step number |
| Time (%) | Cumulative time (%) | Category | Operation | GFlops/sec |
|---|
Number of Hosts used:
TPU type: Cloud TPU
Number of TPU cores:
Batch size:
Show top
ops
off
Top 90%
Time
Name
TensorFlow Op
FLOPS
â–¶
â–¼
FLOPS utilization:
Expression:
Performance information for individual fused operations is not available.
Select items within this category for performance details.
Layout:
Overall TPU FLOPS utilization is
Modifying your model's architecture, data dimensions, and improving the efficiency of CPU operations may help reach the TPU's FLOPS potential.
Settings
DATA
Use categorical coloring
For metadata fields that have many unique values we use a gradient color map
by default. This checkbox allows you to force categorical coloring by a given
metadata field.
Sphereize data
The data is normalized by shifting each point by the centroid and making
it unit norm.
| Checkpoint: | / |
| Metadata: | / |
t-SNE
PCA
Principal component analysis
Custom
Search for two vectors upon which to project all points.
Iteration: 384
For fast results, the data will be sampled down to 10,000 points.
PCA is approximate.
Total variance described: 17.9%.
For fast results, the data was sampled to 50,000 points and randomly projected down to 200 dimensions.
Enable/disable regex mode.
Enable/disable regex mode.
Enable/disable regex mode.
Enable/disable regex mode.
Help with interaction controls.
Enable/disable regex mode.
BOOKMARKS (0)
Showing the first 100,000 of 100,535 data points